{
    "info": {
        "author": "Alex Parinov",
        "author_email": "creafz@gmail.com",
        "bugtrack_url": null,
        "classifiers": [
            "Development Status :: 4 - Beta",
            "Intended Audience :: Developers",
            "License :: OSI Approved :: MIT License",
            "Operating System :: OS Independent",
            "Programming Language :: Python",
            "Programming Language :: Python :: 3",
            "Programming Language :: Python :: 3.5",
            "Programming Language :: Python :: 3.6",
            "Programming Language :: Python :: 3.7"
        ],
        "description": "### Fine-tune pretrained Convolutional Neural Networks with PyTorch.\n\n\n[![PyPI](https://img.shields.io/pypi/v/cnn-finetune.svg)](https://pypi.org/project/cnn-finetune/)\n[![CircleCI](https://circleci.com/gh/creafz/pytorch-cnn-finetune/tree/master.svg?style=shield)](https://circleci.com/gh/creafz/pytorch-cnn-finetune)\n[![codecov.io](https://codecov.io/github/creafz/pytorch-cnn-finetune/coverage.svg?branch=master)](https://codecov.io/github/creafz/pytorch-cnn-finetune)\n\n\n### Features\n- Gives access to the most popular CNN architectures pretrained on ImageNet.\n- Automatically replaces classifier on top of the network, which allows you to train a network with a dataset that has a different number of classes.\n- Allows you to use images with any resolution (and not only the resolution that was used for training the original model on ImageNet).\n- Allows adding a Dropout layer or a custom pooling layer.\n\n\n### Supported architectures and models\n\n#### From the [torchvision](https://github.com/pytorch/vision/) package:\n\n- ResNet (`resnet18`, `resnet34`, `resnet50`, `resnet101`, `resnet152`)\n- ResNeXt (`resnext50_32x4d`, `resnext101_32x8d`)\n- DenseNet (`densenet121`, `densenet169`, `densenet201`, `densenet161`)\n- Inception v3 (`inception_v3`)\n- VGG (`vgg11`, `vgg11_bn`, `vgg13`, `vgg13_bn`, `vgg16`, `vgg16_bn`, `vgg19`, `vgg19_bn`)\n- SqueezeNet (`squeezenet1_0`, `squeezenet1_1`)\n- MobileNet V2 (`mobilenet_v2`)\n- ShuffleNet v2 (`shufflenet_v2_x0_5`, `shufflenet_v2_x1_0`)\n- AlexNet (`alexnet`)\n- GoogLeNet (`googlenet`)\n\n#### From the [Pretrained models for PyTorch](https://github.com/Cadene/pretrained-models.pytorch) package:\n- ResNeXt (`resnext101_32x4d`, `resnext101_64x4d`)\n- NASNet-A Large (`nasnetalarge`)\n- NASNet-A Mobile (`nasnetamobile`)\n- Inception-ResNet v2 (`inceptionresnetv2`)\n- Dual Path Networks (`dpn68`, `dpn68b`, `dpn92`, `dpn98`, `dpn131`, `dpn107`)\n- Inception v4 (`inception_v4`)\n- Xception (`xception`)\n- Squeeze-and-Excitation Networks (`senet154`, `se_resnet50`, `se_resnet101`, `se_resnet152`, `se_resnext50_32x4d`, `se_resnext101_32x4d`)\n- PNASNet-5-Large (`pnasnet5large`)\n- PolyNet (`polynet`)\n\n\n### Requirements\n* Python 3.5+\n* PyTorch 1.1+\n\n### Installation\n\n```\npip install cnn_finetune\n```\n\n\n### Major changes:\n#### Version 0.4\n- Default value for `pretrained` argument in `make_model` is changed from `False` to `True`. Now call `make_model('resnet18', num_classes=10)` is equal to `make_model('resnet18', num_classes=10, pretrained=True)`\n\n\n### Example usage:\n\n#### Make a model with ImageNet weights for 10 classes\n\n```\nfrom cnn_finetune import make_model\n\nmodel = make_model('resnet18', num_classes=10, pretrained=True)\n```\n\n#### Make a model with Dropout\n```\nmodel = make_model('nasnetalarge', num_classes=10, pretrained=True, dropout_p=0.5)\n```\n\n#### Make a model with Global Max Pooling instead of Global Average Pooling\n```\nimport torch.nn as nn\n\nmodel = make_model('inceptionresnetv2', num_classes=10, pretrained=True, pool=nn.AdaptiveMaxPool2d(1))\n```\n\n\n#### Make a VGG16 model that takes images of size 256x256 pixels\nVGG and AlexNet models use fully-connected layers, so you have to additionally pass the input size of images\nwhen constructing a new model. This information is needed to determine the input size of fully-connected layers.\n```\nmodel = make_model('vgg16', num_classes=10, pretrained=True, input_size=(256, 256))\n```\n\n\n#### Make a VGG16 model that takes images of size 256x256 pixels and uses a custom classifier\n```\nimport torch.nn as nn\n\ndef make_classifier(in_features, num_classes):\n    return nn.Sequential(\n        nn.Linear(in_features, 4096),\n        nn.ReLU(inplace=True),\n        nn.Linear(4096, num_classes),\n    )\n\nmodel = make_model('vgg16', num_classes=10, pretrained=True, input_size=(256, 256), classifier_factory=make_classifier)\n```\n\n\n#### Show preprocessing that was used to train the original model on ImageNet\n```\n>> model = make_model('resnext101_64x4d', num_classes=10, pretrained=True)\n>> print(model.original_model_info)\nModelInfo(input_space='RGB', input_size=[3, 224, 224], input_range=[0, 1], mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])\n>> print(model.original_model_info.mean)\n[0.485, 0.456, 0.406]\n```\n\n#### CIFAR10 Example\nSee [examples/cifar10.py](examples/cifar10.py) file (requires PyTorch 1.1+).",
        "description_content_type": "text/markdown",
        "docs_url": null,
        "download_url": "",
        "downloads": {
            "last_day": -1,
            "last_month": -1,
            "last_week": -1
        },
        "home_page": "https://github.com/creafz/pytorch-cnn-finetune",
        "keywords": "",
        "license": "MIT",
        "maintainer": "",
        "maintainer_email": "",
        "name": "cnn-finetune",
        "package_url": "https://pypi.org/project/cnn-finetune/",
        "platform": "",
        "project_url": "https://pypi.org/project/cnn-finetune/",
        "project_urls": {
            "Homepage": "https://github.com/creafz/pytorch-cnn-finetune"
        },
        "release_url": "https://pypi.org/project/cnn-finetune/0.6.0/",
        "requires_dist": null,
        "requires_python": ">=3.5",
        "summary": "Fine-tune pretrained Convolutional Neural Networks with PyTorch",
        "version": "0.6.0"
    },
    "last_serial": 5632820,
    "releases": {
        "0.1.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "277ab986717399d9c6b6c2ba21ae8ecf",
                    "sha256": "fa699a97f7ce95898490c777543bcf36edc17b7ba7ef09ec2b22c166c6c32b0b"
                },
                "downloads": -1,
                "filename": "cnn_finetune-0.1.0.tar.gz",
                "has_sig": false,
                "md5_digest": "277ab986717399d9c6b6c2ba21ae8ecf",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": ">=3.5",
                "size": 8208,
                "upload_time": "2018-02-19T06:34:52",
                "url": "https://files.pythonhosted.org/packages/f8/e7/3c4c6d65cebd7dca364ee1967a16025740c83c252f74d9cb647997a18935/cnn_finetune-0.1.0.tar.gz"
            }
        ],
        "0.1.1": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "8e8b42b1e4b454f1139edff0eadddd42",
                    "sha256": "9faad80625e73a7972e78f35d945a7c0cfc45d457a173f8e8268545d9fb58a8a"
                },
                "downloads": -1,
                "filename": "cnn_finetune-0.1.1.tar.gz",
                "has_sig": false,
                "md5_digest": "8e8b42b1e4b454f1139edff0eadddd42",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": ">=3.5",
                "size": 8203,
                "upload_time": "2018-02-19T06:36:06",
                "url": "https://files.pythonhosted.org/packages/ea/bd/a36374cc3df4e98138b786c1f48dd28f4662378f037a81fea72e5ad17133/cnn_finetune-0.1.1.tar.gz"
            }
        ],
        "0.1.2": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "760c167fff3474e5bd22706471ddd2ff",
                    "sha256": "24b06a33e2f63e32e7fe6140d06873df6895d9c5a3ae4c80f12260dfbe0d5bc4"
                },
                "downloads": -1,
                "filename": "cnn_finetune-0.1.2.tar.gz",
                "has_sig": false,
                "md5_digest": "760c167fff3474e5bd22706471ddd2ff",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": ">=3.5",
                "size": 8220,
                "upload_time": "2018-02-19T06:45:23",
                "url": "https://files.pythonhosted.org/packages/2d/62/403dad1f5ee96eaa9131a6bd4a2823eee94752328ff2063d9eeb69dada1e/cnn_finetune-0.1.2.tar.gz"
            }
        ],
        "0.1.3": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "f4ceb351272df70257c93f86048506a7",
                    "sha256": "d9ed6d0bbdabf284514d253aa88c924751e3d04125c837caf026742b621c8d72"
                },
                "downloads": -1,
                "filename": "cnn_finetune-0.1.3.tar.gz",
                "has_sig": false,
                "md5_digest": "f4ceb351272df70257c93f86048506a7",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": ">=3.5",
                "size": 9208,
                "upload_time": "2018-02-19T07:03:43",
                "url": "https://files.pythonhosted.org/packages/ae/b9/a328bd2d10787d44019bf4897686117ff231eb8699c3a841d7a20f3fc957/cnn_finetune-0.1.3.tar.gz"
            }
        ],
        "0.1.4": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "7522068682a510ab29f9ac967c68cc07",
                    "sha256": "8c71f83b6f42963995fcc7402d51401365682a738db5c53bce982e3b71342b37"
                },
                "downloads": -1,
                "filename": "cnn_finetune-0.1.4.tar.gz",
                "has_sig": false,
                "md5_digest": "7522068682a510ab29f9ac967c68cc07",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": ">=3.5",
                "size": 9582,
                "upload_time": "2018-03-25T13:05:40",
                "url": "https://files.pythonhosted.org/packages/58/11/bd6aaad125bd7cd8af0f89998cb012d3bd04d14e0841bd06e79999060d20/cnn_finetune-0.1.4.tar.gz"
            }
        ],
        "0.1.5": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "623649d4024e9ce586281a7d6a1db556",
                    "sha256": "96ae782a05757ea9b3fbb570ed446f895728157a85b5fd04bfe0db35663abef5"
                },
                "downloads": -1,
                "filename": "cnn_finetune-0.1.5.tar.gz",
                "has_sig": false,
                "md5_digest": "623649d4024e9ce586281a7d6a1db556",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": ">=3.5",
                "size": 9582,
                "upload_time": "2018-03-26T07:10:52",
                "url": "https://files.pythonhosted.org/packages/d2/9b/116d99b7363e7a22f72c3ea02be8aefd8758e96e232411a3ca1a6a10ddda/cnn_finetune-0.1.5.tar.gz"
            }
        ],
        "0.1.6": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "038c0e218a4039877a8989458e05823f",
                    "sha256": "f3c6280773c79d7426e74125fea7ee2a66f62ac622ed91dd51002db00d600137"
                },
                "downloads": -1,
                "filename": "cnn_finetune-0.1.6.tar.gz",
                "has_sig": false,
                "md5_digest": "038c0e218a4039877a8989458e05823f",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": ">=3.5",
                "size": 9592,
                "upload_time": "2018-04-07T05:50:03",
                "url": "https://files.pythonhosted.org/packages/ea/22/7cd61eacb8c3595539136f578bd3963cd73555ed22f5ff16dce11ce945a9/cnn_finetune-0.1.6.tar.gz"
            }
        ],
        "0.2.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "3cb0e4a7d8a243bbd3a2b7e0d75f7eae",
                    "sha256": "a0fbd3f45e2a0ae7b05d0a84bcc08fb8c15f707c3b3fe4fea896cca714685735"
                },
                "downloads": -1,
                "filename": "cnn_finetune-0.2.0.tar.gz",
                "has_sig": false,
                "md5_digest": "3cb0e4a7d8a243bbd3a2b7e0d75f7eae",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": ">=3.5",
                "size": 9751,
                "upload_time": "2018-04-21T05:04:48",
                "url": "https://files.pythonhosted.org/packages/7f/a2/654d5d239bfa20a7f3f0a497c0ffbf554ead8fcd924ae62a434ac1e90f15/cnn_finetune-0.2.0.tar.gz"
            }
        ],
        "0.2.2": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "d5479397849f6d1f25a3f972da5086a7",
                    "sha256": "612ba3f053466e3215e1aad08375175538514140fbc1652598c05252face4ac2"
                },
                "downloads": -1,
                "filename": "cnn_finetune-0.2.2.tar.gz",
                "has_sig": false,
                "md5_digest": "d5479397849f6d1f25a3f972da5086a7",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": ">=3.5",
                "size": 9817,
                "upload_time": "2018-04-21T05:21:17",
                "url": "https://files.pythonhosted.org/packages/ab/01/3e8d53529416a6c6e40b0b61d96ad5d5051c34aae97411a23f88f15421b5/cnn_finetune-0.2.2.tar.gz"
            }
        ],
        "0.3": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "0968c6644ec79a56e68c21a6ff750d63",
                    "sha256": "a2ab89efca11045bf94e929c0dd0e32d1bb21ad3fea0eaa8ec7baf27f436eb4f"
                },
                "downloads": -1,
                "filename": "cnn_finetune-0.3.tar.gz",
                "has_sig": false,
                "md5_digest": "0968c6644ec79a56e68c21a6ff750d63",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": ">=3.5",
                "size": 10262,
                "upload_time": "2018-04-22T10:12:56",
                "url": "https://files.pythonhosted.org/packages/91/d1/d4fc4e37025273f18ccc01e3d808f7228f453d7dbf1e39941973a1b40568/cnn_finetune-0.3.tar.gz"
            }
        ],
        "0.4": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "9795dc36a0501d1c91aa04f2454b188e",
                    "sha256": "bb44b11bbbd78fc0c088c2b5a21e2599010e5aaba78bbb80b50aa5507281ad6e"
                },
                "downloads": -1,
                "filename": "cnn_finetune-0.4.tar.gz",
                "has_sig": false,
                "md5_digest": "9795dc36a0501d1c91aa04f2454b188e",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": ">=3.5",
                "size": 10447,
                "upload_time": "2018-05-20T13:00:17",
                "url": "https://files.pythonhosted.org/packages/05/92/56340e6926f129d6ae305b70da6be7650c0112a44b227b2e98acb0b9403a/cnn_finetune-0.4.tar.gz"
            }
        ],
        "0.5": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "2aaab4d24165fd7f8c75dfda7671113e",
                    "sha256": "f67d74fc6fef040ab3ae10bb2b54352097b3d687f265274a6718b99922a47f25"
                },
                "downloads": -1,
                "filename": "cnn_finetune-0.5.tar.gz",
                "has_sig": false,
                "md5_digest": "2aaab4d24165fd7f8c75dfda7671113e",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": ">=3.5",
                "size": 10743,
                "upload_time": "2018-06-20T15:00:37",
                "url": "https://files.pythonhosted.org/packages/84/95/f1a4420936030c2876728be72f6b1c8b51b9129d231396776331c6c23e8d/cnn_finetune-0.5.tar.gz"
            }
        ],
        "0.5.1": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "5cf578480da8558ae61a054d05a68359",
                    "sha256": "e0cf63185972cffde8a2778140f999aa25cc4e83d4fab98d429a8f844271374e"
                },
                "downloads": -1,
                "filename": "cnn_finetune-0.5.1.tar.gz",
                "has_sig": false,
                "md5_digest": "5cf578480da8558ae61a054d05a68359",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": ">=3.5",
                "size": 11401,
                "upload_time": "2018-08-22T15:29:52",
                "url": "https://files.pythonhosted.org/packages/21/c7/97f7056e0e82f3d25cb9049b133394d52ac3008be6887d3894b23465e7b3/cnn_finetune-0.5.1.tar.gz"
            }
        ],
        "0.5.2": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "6315c27dc0edc1201a843554866ee8bb",
                    "sha256": "6fd65d6889a9d30a288cd7f28eb3daa2a6078d0cab06e85837f5921eb3926769"
                },
                "downloads": -1,
                "filename": "cnn_finetune-0.5.2.tar.gz",
                "has_sig": false,
                "md5_digest": "6315c27dc0edc1201a843554866ee8bb",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 10820,
                "upload_time": "2018-12-20T07:48:13",
                "url": "https://files.pythonhosted.org/packages/f5/11/a8a7d83d327ad4aafe6ab7ef8a533ff00dccc4300e2becd05fcdf4f38455/cnn_finetune-0.5.2.tar.gz"
            }
        ],
        "0.5.3": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "19a5885182976ec46e7c1441fc8eb135",
                    "sha256": "c95075b5a5b460589d2fd7ece7ee7962a12dd266f8ecbe9b51b51f958419788b"
                },
                "downloads": -1,
                "filename": "cnn_finetune-0.5.3.tar.gz",
                "has_sig": false,
                "md5_digest": "19a5885182976ec46e7c1441fc8eb135",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": ">=3.5",
                "size": 10911,
                "upload_time": "2018-12-20T07:53:31",
                "url": "https://files.pythonhosted.org/packages/a6/8f/0f7078b376eaeb331cf99eb94a43cf700556cbcb146f1e4f042d21450aad/cnn_finetune-0.5.3.tar.gz"
            }
        ],
        "0.6.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "35b49ffbcfc63da9218576c2a76ecb84",
                    "sha256": "557582afa0acfbdc93c4d18a34db0b34714086c1194f9731056fcb425a5937bd"
                },
                "downloads": -1,
                "filename": "cnn_finetune-0.6.0.tar.gz",
                "has_sig": false,
                "md5_digest": "35b49ffbcfc63da9218576c2a76ecb84",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": ">=3.5",
                "size": 11181,
                "upload_time": "2019-08-05T06:39:45",
                "url": "https://files.pythonhosted.org/packages/e4/63/03a442d31401c43fc17a814f22bd7c39ab8f13f42a6b2467ca0d0d042b3a/cnn_finetune-0.6.0.tar.gz"
            }
        ]
    },
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "md5": "35b49ffbcfc63da9218576c2a76ecb84",
                "sha256": "557582afa0acfbdc93c4d18a34db0b34714086c1194f9731056fcb425a5937bd"
            },
            "downloads": -1,
            "filename": "cnn_finetune-0.6.0.tar.gz",
            "has_sig": false,
            "md5_digest": "35b49ffbcfc63da9218576c2a76ecb84",
            "packagetype": "sdist",
            "python_version": "source",
            "requires_python": ">=3.5",
            "size": 11181,
            "upload_time": "2019-08-05T06:39:45",
            "url": "https://files.pythonhosted.org/packages/e4/63/03a442d31401c43fc17a814f22bd7c39ab8f13f42a6b2467ca0d0d042b3a/cnn_finetune-0.6.0.tar.gz"
        }
    ]
}